Malware Classification Using Artificial Neural Network
R K Charan, Chavan Chandu Nayak, K N V Khasim, Ashwini Kodipalli, A Ushasree., B. Rajasekar · 2025
The rapid increase of malware poses a significant cybersecurity threat, necessitating effective detection and classification techniques. Traditional signature-based methods often fall short due to their inability to recognize novel and evolving malware strains. This paper explores the application of Artificial Neural Networks (ANN) for malware classification, leveraging their capability to learn complex patterns and generalize from training data. We developed and fine-tuned a machine learning algorithm based on neural networks. We utilized a diverse and extensive collection of data, which included samples of both harmful malware and legitimate software applications. The proposed ANN model achieved a high classification accuracy, demonstrating its efficacy in distinguishing between malicious and non-malicious executables. Comparative analysis of different optimizer techniques highlights the superior performance of our ANN based approach. The results suggest that ANNs can significantly enhance malware detection systems, offering a robust solution for cybersecurity defenses. Future work will focus on optimizing the model and expanding the dataset to further improve classification accuracy and adaptability to emerging threats.